Adolescent depression is a substantial public health problem which contributes to academic failure, poor social function, substance use disorders, teen pregnancy, chronic morbidity, and completed suicides. Existing treatment approaches for adolescent depression such as psychotherapy and antidepressant medications are often ineffective and do not target relevant neurobiology. Repetitive transcranial magnetic stimulation (rTMS) applied to the left dorsolateral prefrontal cortex (L-DLPFC) is a Food and Drug Administration approved treatment of major depressive disorder in adults. In clinical practice 10 Hz (high frequency) rTMS is typically used to treat depression but prior studies suggest that 1 Hz (low frequency) rTMS may be as effective, more tolerable, and safer compared to 10 Hz rTMS. Initial studies of rTMS for adolescent depression are promising but the optimal dosing and brain effects of rTMS treatments may be different than adults. One theory is that treatment with rTMS impacts gamma-aminobutyric acid (GABA) and glutamate neurotransmitters. However, there are no biomarker approaches or clinical profiles to guide rTMS treatments in adolescents. To solve these problems we propose a novel dose finding and target validation study of rTMS for adolescent depression with a biomarker stratified randomized design. Depressed adolescents will be stratified according to a baseline measure of cortical glutamatergic tone called intracortical facilitation (ICF) and then randomized to six weeks of 1 Hz or 10 Hz rTMS treatment applied to the L-DLPFC. We will determine if baseline measures of ICF predict antidepressant response to 1 Hz or 10 Hz rTMS in depressed adolescents and if ICF measures change over time with successful rTMS treatment. The research team will study and develop biomarkers of GABA and glutamate tone with ultrahigh-field, 7 tesla (7 T) magnetic resonance spectroscopy (MRS) measures of the bilateral dorsolateral prefrontal cortices. We will also examine the effect of baseline anhedonia and historical adversity on treatment response to rTMS with the aim of understanding these clinical characteristics and developing clinical treatment targets with links to GABA and glutamate tone. We hypothesize that depressed adolescents with high baseline ICF measures will have a greater antidepressant response (assessed by greater decrease in Children?s Depression Rating Scale-Revised scores) to 1 Hz rTMS while depressed adolescents with low baseline ICF will have a greater antidepressant response to 10 Hz rTMS. We also hypothesize that 1 Hz rTMS will increase cortical GABA and decrease cortical glutamate, while 10 Hz rTMS will decrease cortical GABA and increase cortical glutamate (measured with bilateral 7 T MRS of the dorsolateral prefrontal cortices). The proposed research program will answer important questions regarding the neurobiology of adolescent depression, the mechanisms of rTMS treatments in adolescents, and dosing rTMS in adolescent depression. Findings will inform larger studies of rTMS and the development of novel treatments for adolescent depression.